Matches in SemOpenAlex for { <https://semopenalex.org/work/W3108436304> ?p ?o ?g. }
- W3108436304 endingPage "482" @default.
- W3108436304 startingPage "471" @default.
- W3108436304 abstract "Abstract Recent GAN‐based image inpainting approaches adopt an average strategy to discriminate the generated image and output a scalar, which inevitably lose the position information of visual artifacts. Moreover, the adversarial loss and reconstruction loss (e.g., ℓ 1 loss) are combined with tradeoff weights, which are also difficult to tune. In this paper, we propose a novel detection‐based generative framework for image inpainting, which adopts the min‐max strategy in an adversarial process. The generator follows an encoder‐decoder architecture to fill the missing regions, and the detector using weakly supervised learning localizes the position of artifacts in a pixel‐wise manner. Such position information makes the generator pay attention to artifacts and further enhance them. More importantly, we explicitly insert the output of the detector into the reconstruction loss with a weighting criterion, which balances the weight of the adversarial loss and reconstruction loss automatically rather than manual operation. Experiments on multiple public datasets show the superior performance of the proposed framework. The source code is available at https://github.com/Evergrow/GDN_Inpainting." @default.
- W3108436304 created "2020-12-07" @default.
- W3108436304 creator A5003969300 @default.
- W3108436304 creator A5017215030 @default.
- W3108436304 creator A5034818978 @default.
- W3108436304 creator A5045242540 @default.
- W3108436304 creator A5068027800 @default.
- W3108436304 date "2020-10-01" @default.
- W3108436304 modified "2023-09-26" @default.
- W3108436304 title "Pixel‐wise Dense Detector for Image Inpainting" @default.
- W3108436304 cites W1834627138 @default.
- W3108436304 cites W1903029394 @default.
- W3108436304 cites W1975049209 @default.
- W3108436304 cites W1993120651 @default.
- W3108436304 cites W2055132753 @default.
- W3108436304 cites W2057016804 @default.
- W3108436304 cites W2100415658 @default.
- W3108436304 cites W2108598243 @default.
- W3108436304 cites W2115273023 @default.
- W3108436304 cites W2133665775 @default.
- W3108436304 cites W2145023731 @default.
- W3108436304 cites W2183341477 @default.
- W3108436304 cites W2331128040 @default.
- W3108436304 cites W2475287302 @default.
- W3108436304 cites W2557414982 @default.
- W3108436304 cites W2572730214 @default.
- W3108436304 cites W2732026016 @default.
- W3108436304 cites W2738588019 @default.
- W3108436304 cites W2963073614 @default.
- W3108436304 cites W2963231084 @default.
- W3108436304 cites W2963270367 @default.
- W3108436304 cites W2969342617 @default.
- W3108436304 cites W2981682056 @default.
- W3108436304 cites W2985764327 @default.
- W3108436304 cites W2989207674 @default.
- W3108436304 cites W3035002246 @default.
- W3108436304 cites W3100341797 @default.
- W3108436304 cites W4249962038 @default.
- W3108436304 doi "https://doi.org/10.1111/cgf.14160" @default.
- W3108436304 hasPublicationYear "2020" @default.
- W3108436304 type Work @default.
- W3108436304 sameAs 3108436304 @default.
- W3108436304 citedByCount "8" @default.
- W3108436304 countsByYear W31084363042021 @default.
- W3108436304 countsByYear W31084363042022 @default.
- W3108436304 countsByYear W31084363042023 @default.
- W3108436304 crossrefType "journal-article" @default.
- W3108436304 hasAuthorship W3108436304A5003969300 @default.
- W3108436304 hasAuthorship W3108436304A5017215030 @default.
- W3108436304 hasAuthorship W3108436304A5034818978 @default.
- W3108436304 hasAuthorship W3108436304A5045242540 @default.
- W3108436304 hasAuthorship W3108436304A5068027800 @default.
- W3108436304 hasBestOaLocation W31084363042 @default.
- W3108436304 hasConcept C10138342 @default.
- W3108436304 hasConcept C111919701 @default.
- W3108436304 hasConcept C115961682 @default.
- W3108436304 hasConcept C11727466 @default.
- W3108436304 hasConcept C118505674 @default.
- W3108436304 hasConcept C121332964 @default.
- W3108436304 hasConcept C126838900 @default.
- W3108436304 hasConcept C153180895 @default.
- W3108436304 hasConcept C154945302 @default.
- W3108436304 hasConcept C160633673 @default.
- W3108436304 hasConcept C162324750 @default.
- W3108436304 hasConcept C163258240 @default.
- W3108436304 hasConcept C177264268 @default.
- W3108436304 hasConcept C183115368 @default.
- W3108436304 hasConcept C198082294 @default.
- W3108436304 hasConcept C199360897 @default.
- W3108436304 hasConcept C200873422 @default.
- W3108436304 hasConcept C2776760102 @default.
- W3108436304 hasConcept C2780992000 @default.
- W3108436304 hasConcept C2988416141 @default.
- W3108436304 hasConcept C31972630 @default.
- W3108436304 hasConcept C41008148 @default.
- W3108436304 hasConcept C62520636 @default.
- W3108436304 hasConcept C71924100 @default.
- W3108436304 hasConcept C76155785 @default.
- W3108436304 hasConcept C94915269 @default.
- W3108436304 hasConcept C98045186 @default.
- W3108436304 hasConceptScore W3108436304C10138342 @default.
- W3108436304 hasConceptScore W3108436304C111919701 @default.
- W3108436304 hasConceptScore W3108436304C115961682 @default.
- W3108436304 hasConceptScore W3108436304C11727466 @default.
- W3108436304 hasConceptScore W3108436304C118505674 @default.
- W3108436304 hasConceptScore W3108436304C121332964 @default.
- W3108436304 hasConceptScore W3108436304C126838900 @default.
- W3108436304 hasConceptScore W3108436304C153180895 @default.
- W3108436304 hasConceptScore W3108436304C154945302 @default.
- W3108436304 hasConceptScore W3108436304C160633673 @default.
- W3108436304 hasConceptScore W3108436304C162324750 @default.
- W3108436304 hasConceptScore W3108436304C163258240 @default.
- W3108436304 hasConceptScore W3108436304C177264268 @default.
- W3108436304 hasConceptScore W3108436304C183115368 @default.
- W3108436304 hasConceptScore W3108436304C198082294 @default.
- W3108436304 hasConceptScore W3108436304C199360897 @default.
- W3108436304 hasConceptScore W3108436304C200873422 @default.
- W3108436304 hasConceptScore W3108436304C2776760102 @default.